Show HN: HyperSAE – Hyperbolic Sparse Autoencoders for LLM Interpretability | Dark Hacker News

Dark Hacker News

new|best|ask|show|jobs

Built with the HackerNews API. Developed by LJT.AI

Note: The HackerNews API provides up to 500 stories per category

Show HN: HyperSAE – Hyperbolic Sparse Autoencoders for LLM Interpretability | Dark Hacker News

Show HN: HyperSAE – Hyperbolic Sparse Autoencoders for LLM Interpretability(github.com)

3 points by visha1v 68 days ago | 0 comments

No comments yet